Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    names or words with the letter 'h' as the penultimate character

    oai_token-act-pair · gpt-3.5-turbo

    instances of certain letters or letter combinations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     takeoff
    -0.65
     composing
    -0.63
     compliment
    -0.63
     bordering
    -0.59
     behold
    -0.58
     underrated
    -0.58
    ipolar
    -0.57
    theless
    -0.57
     hither
    -0.57
     ACPI
    -0.56
    POSITIVE LOGITS
    ateg
    0.80
    olars
    0.77
    irts
    0.77
    uler
    0.77
    raper
    0.75
    apers
    0.75
    nikov
    0.74
    Ó
    0.73
    okes
    0.72
    edIn
    0.72
    Act Density 0.191%

    No Known Activations